Multi-informant evaluation of students' psychosocial status through SDQ in a national Greek sample

Abstract

Background

This paper presents evidence from a European funded program with national contribution regarding the screening of students using the Strengths and Difficulties Questionnaire (SDQ). The aim of the present study is two-fold: a) to provide a thorough investigation of the factor structure of the Greek parent, teacher, and student versions of the SDQ, and b) to examine psychosocial functioning of a non-clinical student sample, as well as potential differences as a function of student gender, age, grade or year of schooling, and immigration background

Method

The study instruments (SDQ, The Short Mood and Feelings Questionnaire, and The Classmate Support Scale) were administered to a national sample of 6290 students, aged 11 to 15 years.

Results

Results provided support for the hypothesised five-factor structure of the SDQ. Agreement was higher between parent and student reports. Immigrants had higher scores in total difficulties, hyperactivity-inattention, and emotional problems than natives, whereas natives were given higher scores on peer problems and prosocial behaviour.

Conclusions

Assessing the psychosocial well being of students' based on a multi-informant approach has implications for schools and mental health professionals.

Introduction

The mental health of children and adolescents is of prime importance worldwide. The use of epidemiological studies has been reported as necessary for the accurate assessement trends in population prevalence (Achenbach et al., 2008. Braddick, Carral, Jenkins, & Jane-Llopis, 2009. Goodman, 2001). To our knowledge, in Greece there are few epidemilogical studies assessing trends and giving information on children's/adolescents' mental health (Giannakopoulos et al., 2013; Kolaitis, Fyssas, & Tsiantis, 2009), resulting in inadequate data, both in terms of quality and quantity. In addition, the absence of administrative process data from the health system, the malfunctioning of health facilities at primary care and specialist care levels, as well as the absence of school psychology services, form the main barriers in methodically gathering data overtime. Moreover, the drastic funding cuts to Greek child and adolescent mental health services since 2007 as a result of the financial crisis has meant a reduction in care and treatment services (Kolaitis et al., 2009).

The present paper presents data from a European funded program regarding the screening of primary and secondary students using the Strengths and Difficulties Questionnaire (SDQ). This project is part of a larger national four-year pilot programme implementing psychological services to public schools that have at least 30% migrant students (Diapolis Programme, 2010-2013). The aims of the national program were, among others, the development of a pilot school support system with specialized psychological support services for students and their families, and the implementation of short-term multilevel prevention and intervention initiatives targeting all interested groups (students, teachers, parents). Part of the needs assessment stage was the administration of SDQ to all participant groups.

In many countries, social services and educational authorities have promoted the use of strengths-based measures as part of the assessment process in schools, social services, and child and adolescent mental health agencies (Sointu, Savolainen, & Lappalainen, 2012). The SDQ is one of the most common screening tools used in clinical and educational settings for the assessment of children's psychosocial status (Chiorri, Hall, Casely-Hayford, & Malmberg, 2016). Multiple perspective assessment is valuable since it is grounded on the fact that the recognition of psychosocial problems tends to be situational and highly determined by children's relationships and informants' different mind-sets for judging and reporting child behaviour (Boman et al., 2016). There appears to be a partial conflicting correlation between teachers'and parents'SDQ ratings, highly and extensively searched in the literature (e.g. see De Los Reyes & Kazdin, 2005 for a theoretical model interpreting informant discrepancies). The parent and teacher reports are more often used in research, even though there is evidence to suggest that in the case of emotional difficulties self-report information is just as useful as teacher data, but less useful than parent data.

Internationally, there is conflicting evidence regarding the factor stucture of the SDQ (Boman et al., 2016; McCrory & Layte, 2012), as well as the inter-rater reliability of the instrument (Achenbach et al., 2008). Previous research studies conducted in diverse cultural contexts generally provided support for the hypothesised five-factor structure of the SDQ (Bøe, Hysing, Skogen, & Breivik, 2016, Giannakopoulos et al., 2010. Hill & Hughes, 2007. Ruchkin, Koposov, & Schwab-Stone, 2007. Vogels, Crone, Hoekstra, & Reijneveld, 2009), whereas others suggested that the SDQ does not have a very “clean” factor structure (Mellor & Stokes, 2007. Ronning, Handergaard, Sourander, & Morch, 2004). Alternative theoretically justifiable models have been validated elsewhere in the literature. Consistent with Goodman's original conceptualisation of the SDQ, Yao et al. (2009) found that a hierarchical model with a total difficulties factor underlying the four problem scales provided acceptable model fit among adolescents. Other investigators have tested a model where emotional problems and peer problems are replaced by an internalising problems factor, and conduct problems and hyperactivity-inatention scales with an externalising problems factor, whilst still retaining a separate prosocial behaviour factor. This three-factor model provided acceptable model fit for children in Italy (Di Riso et al., 2010), and in Belgium (van Leeuwen, Meerschaert, Bosmans, De Medts, & Braet, 2006). In the UK, Goodman, Lamping, and Ploubidis (2010) provided little support for replacing the scales with internalising and externalising factors. Instead, they found that a hierarchical model with higher-order internalising and externalising factors (and a separate prosocial behaviour factor) achieved acceptable model fit indices for the self-report versions of the scale. Finally, the evidence is also conflicting regarding the migrant students' risk of mental health problems (Boman et al., 2016). In Greece, adolescents' emotional and behavioural problems were documented by Giannakopoulos et al. (2010) and Giannakopoulos et al. (2013), based on a representative nationwide sample of 1.194 adolescents. Specifically, in the first study, SDQ was completed by adolescents, supporting the five-factor structure of the instrument; in the second study, the completion of SDQ by both adolescents and parents provided evidence of the psychometric characteristics of SDQ. Further, using the same sample, Giannakopoulos, Mihas, Dimitrakaki, and Tountas (2009), documented a series of family factors as potential determinants of adolescent mental health status.

This paper builds on former national and international studies investigating psychometric characteristics of the SDQ and aims: a) to provide an investigation of the factor structure of the Greek parent, teacher, and student versions of the SDQ, and b) to examine psychosocial functioning of a non-clinical student sample through multi-informant evaluation, as well as potential differences as a function of student gender, age, and immigration background. The data was collected around a time of wider recognition of the global economic downturn hitting Greece. Although the present study did not address the socio-economic situation of the families, we believe that this is a serious attempt to map out the psychosocial well being of school students through a multi-informant approach.